GPT-5.6 Luna

Available

A lower-priced GPT-5.6 API option with configurable reasoning and image input.

Access details: OpenAI · Responses API

An alias. The provider can point it at a newer version.

gpt-5.6-luna

Updated Oct 20, 2026

Sources & notes

Capabilities & limits

Context window: 1.05M
Maximum output: 128K
Image input: Supported
Tool calling: Supported
Structured output: Supported

Pricing

USD / 1M tokens

Input: $0.20
Output: $1.20

Standard text rates after the July 30 price change. Above 272K input tokens, different rates apply. Cache, batch, tools, and regional premiums are excluded.

Full rate card 8 rates

All prices in USD.

Short-context input

$0.20 per 1M tokens

prompt ≤ 272K input tokens

Short-context output

$1.20 per 1M tokens

prompt ≤ 272K input tokens

Cached input

$0.02 per 1M tokens

prompt ≤ 272K input tokens

Cache writes

$0.25 per 1M tokens

prompt ≤ 272K input tokens

Long-context input

$0.40 per 1M tokens

prompt > 272K input tokens
rate applies to the full request

Long-context output

$1.80 per 1M tokens

prompt > 272K input tokens
rate applies to the full request

Long-context cached input

$0.04 per 1M tokens

prompt > 272K input tokens
rate applies to the full request

Long-context cache writes

$0.50 per 1M tokens

prompt > 272K input tokens
rate applies to the full request

Other tiers per 1M tokens (short context): Batch: $0.1 input / $0.6 output / $0.01 cached; Flex: $0.1 input / $0.6 output / $0.01 cached; Fast mode: $0.4 input / $2.4 output / $0.04 cached. Regional processing (data residency) endpoints add a 10% uplift. Cache writes are billed at 1.25x the uncached input rate. Luna prices were cut 80% on July 30, 2026.

Price history

Jul 9, 2026

Input — per million tokens

Jul 30, 2026

Input $0.20 per million tokens

Sep 23, 2026

Input $0.20 per million tokens

Jul 9, 2026

Output — per million tokens

Jul 30, 2026

Output $1.20 per million tokens

Sep 23, 2026

Output $1.20 per million tokens

Technical details

Parameters: Not published
Active parameters: Not published
Architecture: Not published
Training / knowledge cutoff: Feb 16, 2026
Input modalities: Text, image
Output modalities: Text

From OpenAI

Official announcement ↗

How they describe it

GPT-5.6 model optimized for cost-sensitive workloads

Release & lifecycle

First available: Jul 9, 2026
available: As of Sep 13, 2026
available: As of Oct 20, 2026

Read the release story →